C-ReD: A Comprehensive Chinese Benchmark for AI-Generated Text Detection Derived from Real-World Prompts

2026-04-13

AI Summary

This work addresses key challenges in Chinese AI-generated text detection—namely, insufficient model diversity, data homogeneity, and unrealistic prompts—by introducing C-ReD, the first benchmark for Chinese AI-generated text detection built upon authentic user prompts. C-ReD encompasses outputs from multiple large language models, spans diverse domains, and features highly varied prompts to better reflect real-world usage. Through a structured evaluation framework, C-ReD substantially enhances the generalization capability of detectors not only on in-domain tasks but also in cross-model and cross-dataset settings, effectively addressing critical gaps in existing Chinese detection benchmarks.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Recently, large language models (LLMs) are capable of generating highly fluent textual content. While they offer significant convenience to humans, they also introduce various risks, like phishing and academic dishonesty. Numerous research efforts have been dedicated to developing algorithms for detecting AI-generated text and constructing relevant datasets. However, in the domain of Chinese corpora, challenges remain, including limited model diversity and data homogeneity. To address these issues, we propose C-ReD: a comprehensive Chinese Real-prompt AI-generated Detection benchmark. Experiments demonstrate that C-ReD not only enables reliable in-domain detection but also supports strong generalization to unseen LLMs and external Chinese datasets-addressing critical gaps in model diversity, domain coverage, and prompt realism that have limited prior Chinese detection benchmarks. We release our resources at https://github.com/HeraldofLight/C-ReD.

Problem

• AI-generated text detection
• Chinese benchmark
• model diversity
• data homogeneity
• prompt realism

Innovation

• AI-generated text detection
• Chinese benchmark
• real-world prompts
• model diversity
• generalization